Perception method, device and system

Abstract

A sensing method, device and system, in the method, a first sensor obtains first data, the first sensor supports different reporting modes, and the content of the first data is related to the reporting modes; sending the first data to a target node; wherein the first data comprises at least one of the following sensing data: original sensing data; receiving / transmitting the channel data; point cloud level data; feature level data; and target-level data. The method is used for enabling the sensor to provide more diversified sensing information to meet demands of different scenes, so that the processing efficiency is improved, and different intelligent driving requirements in different scenes are met.

Description

Technical Field

[0001] This application relates to the field of sensor technology, and in particular to a sensing method, apparatus and system. Background Technology

[0002] The development of sensor technology is very rapid. For example, the detection range of 4D high resolution radar can reach 300 to 350 meters. It can accurately map the 4D environment and achieve a high resolution close to that of lidar. At the same time, it can form a clear point cloud map of the surrounding environment and provide a more comprehensive and clear perception result.

[0003] Taking sensor installations on vehicles as an example, the sensor's perception results can assist vehicle driving or help the vehicle achieve intelligent driving. With the development of new energy vehicles and advanced driving assistance systems (ADAS), the demand for sensor data transmission is also increasing accordingly. Summary of the Invention

[0004] This application provides a perception method, apparatus, and system that enable sensors to provide more diverse perception information to meet the demands of different scenarios, thereby improving processing efficiency and meeting the needs of intelligent driving in different scenarios.

[0005] In a first aspect, this application provides a sensing method that can be applied to a first sensor. The method may include: acquiring first data, wherein the first sensor supports different reporting modes, and the content of the first data is related to the reporting mode; and sending the first data to a target node. The first data includes at least one of the following sensing data: raw sensing data; transmit / receive channel data; point cloud-level data; feature-level data; and target-level data.

[0006] Using the above method, the first sensor can report one or more types of perception data to the target node, providing more diverse perception data to meet the fusion requirements. This further ensures that the target node (e.g., other sensors with computing capabilities or the vehicle's intelligent driving system) performs accurate perception processing based on at least one type of received perception data, thereby improving processing efficiency and meeting the needs of intelligent driving in different scenarios. When the first sensor reports a combination of multiple perception data to the target node, the perception accuracy can also be improved. When the target node is implemented as the vehicle's intelligent driving system, the improved perception accuracy can also ensure that the intelligent driving system makes accurate control decisions, thereby ensuring the safety of vehicle driving.

[0007] In conjunction with the first aspect, in one possible implementation, the first data includes sensing data of different regions of interest (ROIs) within the detection area of ​​the first sensor, wherein the different ROIs correspond to different reporting modes.

[0008] Using the above method, the first sensor can report one or more of various sensing data to the target node based on the granularity of ROI division, enabling the target node to process data for different sensing requirements of different ROIs, thereby improving processing efficiency.

[0009] In conjunction with the first aspect, in one possible implementation, the reporting mode of the first data is related to the scene in which the first sensor is located.

[0010] For example, the first sensor is mounted on the vehicle, and the scenario in which the vehicle is located may be related to the external environmental requirements of the vehicle's location, the vehicle's driving parameters, etc. For example, the scenario in which the first sensor is located may include any of the following scenarios based on the vehicle's driving speed: low-speed driving scenario, medium-speed driving scenario, high-speed driving scenario; or, the scenario in which the first sensor is located may include any of the following scenarios based on the road type in which the vehicle is located: urban road scenario, highway road scenario; or, the scenario in which the first sensor is located may include any of the following scenarios based on the weather conditions in which the vehicle is located: good visibility scenario, moderate visibility scenario, poor visibility scenario; or, the scenario in which the first sensor is located may include any of the following scenarios based on the road conditions of the road segment in which the vehicle is located: severely congested road segment scenario, moderately congested road segment scenario, lightly congested road segment scenario, unobstructed road segment scenario.

[0011] Using the above method, the first sensor can provide diverse perception data to the target node that meets the needs of intelligent driving control decisions, based on the different requirements of perception data in different scenarios. Optionally, the scenarios may also include other scenarios not shown, or different scenarios may be divided according to the needs of autonomous driving functions; this application embodiment does not limit this.

[0012] In conjunction with the first aspect, in one possible implementation, the reporting mode of the first data includes a combination of at least one of the following reporting modes: raw sensing data reporting mode; transmit / receive channel data reporting mode; point cloud-level data reporting mode; feature-level data reporting mode; and target-level data reporting mode. It should be understood that this is merely an illustrative example of the reporting mode for the first data and not a limitation thereof. In specific implementations, depending on the sensing capabilities or supported computing power of the first sensor itself, the reporting mode of the first data can be transformed into other modes, which will not be elaborated upon here.

[0013] In conjunction with the first aspect, in one possible implementation, different types of sensing data satisfy the following priority levels: the priority of the raw sensing data is higher than the priority of the transmit / receive channel data; the priority of the transmit / receive channel data is higher than the priority of the point cloud-level data; and the priority of the point cloud-level data is higher than the priority of the feature-level data and the target-level data.

[0014] By prioritizing different types of sensing data, the first sensor can provide the target node with at least one type of sensing data that better meets the needs of different scenarios or business requirements.

[0015] In conjunction with the first aspect, in one possible implementation, the reporting mode corresponding to different ROIs is related to the priority of the perceived data.

[0016] In conjunction with the first aspect, in one possible implementation, in any of the following scenarios: low-speed driving scenario, urban road scenario, good visibility scenario, severely congested road segment scenario, or moderately congested road segment scenario, the content of the first data includes target-level data of different ROIs of the first sensor; in any of the following scenarios: medium-speed driving scenario, high-speed driving scenario, highway road scenario, general visibility scenario, poor visibility scenario, lightly congested road segment scenario, or unobstructed road segment scenario, the content of the first data includes at least one of the following: raw sensing data of the first ROI, transmit / receive channel data, or point cloud-level data, and target-level data of the second ROI, wherein the detection distance corresponding to the first ROI is less than the detection distance corresponding to the second ROI.

[0017] In conjunction with the first aspect, in one possible implementation, the target node can also issue customized requirements to the first sensor. For example, the method may further include: receiving first indication information from the target node; determining the scene in which the first sensor is located based on the first indication information; wherein the first indication information includes at least one of the following: the speed of the vehicle; the speeds of other vehicles near the vehicle; the geographical type of the detection area; meteorological environmental information corresponding to the detection area; and the load information of the vehicle's computing node. Alternatively, the method may further include: receiving second indication information from the target node, the second indication information indicating the reporting mode of the first sensor or indicating the scene in which the first sensor is located.

[0018] Using the above method, the first sensor can provide more diverse sensing data to the target node according to the target node's customized needs.

[0019] In conjunction with the first aspect, in one possible implementation, the method may further include: determining different ROIs within the detection area of ​​the first sensor based on the field of view or detection distance range of the first sensor; or, receiving third indication information from the target node, the third indication information indicating different ROIs within the detection area of ​​the first sensor.

[0020] In conjunction with the first aspect, in one possible implementation, the target node includes a second sensor; or, the first sensor is mounted on the vehicle, and the target node includes the vehicle's computing node.

[0021] In conjunction with the first aspect, in one possible implementation, the second sensor is located on the same vehicle as the first sensor.

[0022] Secondly, this application provides a perception method that can be applied to a target node. The method may include: receiving first data from a first sensor, wherein the first sensor supports different reporting modes, and the content of the first data is related to the reporting mode; performing perception processing based on the first data; wherein the first data includes at least one of the following perception data: raw perception data; transmit / receive channel data; point cloud-level data; feature-level data; target-level data.

[0023] In conjunction with the second aspect, in one possible implementation, the first data includes sensing data of different regions of interest (ROIs) within the detection area of ​​the first sensor, wherein the different ROIs correspond to different reporting modes.

[0024] In conjunction with the second aspect, in one possible implementation, the reporting mode of the first data is related to the scene in which the first sensor is located.

[0025] In conjunction with the second aspect, in one possible implementation, the first sensor is mounted on a vehicle, and the target node includes either the vehicle's computing node or a second sensor associated with the vehicle. The method further includes: sending first indication information to the first sensor, wherein the first indication information is used to determine the scene in which the first sensor is located, and the first indication information includes at least one of the following: the speed of the vehicle; the speeds of other vehicles near the vehicle; the geographical type of the detection area; the meteorological environment information corresponding to the detection area; and the load information of the vehicle's computing node.

[0026] In conjunction with the second aspect, in one possible implementation, the scenario in which the first sensor is located includes any of the following scenarios based on the vehicle's driving speed: low-speed driving scenario, medium-speed driving scenario, and high-speed driving scenario; or, the scenario in which the first sensor is located includes any of the following scenarios based on the road type in which the vehicle is located: urban road scenario and highway road scenario; or, the scenario in which the first sensor is located includes any of the following scenarios based on the weather conditions at the vehicle's location: good visibility scenario, moderate visibility scenario, and poor visibility scenario; or, the scenario in which the first sensor is located includes any of the following scenarios based on the road conditions of the road segment in which the vehicle is located: severely congested road segment scenario, moderately congested road segment scenario, lightly congested road segment scenario, and unobstructed road segment scenario.

[0027] In conjunction with the second aspect, in one possible implementation, the reporting mode of the first data includes a combination of at least one of the following reporting modes: raw sensing data reporting mode; transmit / receive channel data reporting mode; point cloud level data reporting mode; feature level data reporting mode; target level data reporting mode.

[0028] In conjunction with the second aspect, in one possible implementation, different types of sensing data satisfy the following priority levels: the priority of the raw sensing data is higher than the priority of the transmit / receive channel data; the priority of the transmit / receive channel data is higher than the priority of the point cloud-level data; and the priority of the point cloud-level data is higher than the priority of the feature-level data and the target-level data.

[0029] In conjunction with the second aspect, in one possible implementation, the reporting mode corresponding to different ROIs is related to the priority of the perceived data.

[0030] In conjunction with the second aspect, in one possible implementation, in any of the following scenarios: low-speed driving scenario, urban road scenario, good visibility scenario, severely congested road segment scenario, or moderately congested road segment scenario, the content of the first data includes target-level data of different ROIs of the first sensor; in any of the following scenarios: medium-speed driving scenario, high-speed driving scenario, highway road scenario, general visibility scenario, poor visibility scenario, lightly congested road segment scenario, or unobstructed road segment scenario, the content of the first data includes at least one of the following: raw sensing data of the first ROI, transmit / receive channel data, or point cloud-level data, and target-level data of the second ROI, wherein the detection distance corresponding to the first ROI is less than the detection distance corresponding to the second ROI.

[0031] In conjunction with the second aspect, in one possible implementation, the first sensor is mounted on a vehicle, the target node includes the vehicle's computing node, and the method further includes: sending second indication information to the first sensor, the second indication information indicating the reporting mode of the first sensor or indicating the scene in which the first sensor is located.

[0032] In conjunction with the second aspect, in one possible implementation, the method further includes: determining different ROIs within the detection area of ​​the first sensor based on the field of view or detection distance range of the first sensor; or sending third indication information to the target node, the third indication information indicating different ROIs within the detection area of ​​the first sensor.

[0033] Thirdly, this application provides a communication device including at least one processor and an interface circuit. The interface circuit is used to provide data or code instructions to the at least one processor. The at least one processor is used to implement the method described in the first aspect and any possible implementation of the first aspect through logic circuits or executing code instructions, or to implement the method described in the second aspect and any possible implementation of the second aspect.

[0034] Fourthly, this application provides a sensing device including units for implementing the method as described in the first aspect and any possible implementation of the first aspect, or including units for implementing the method as described in the second aspect and any possible implementation of the second aspect.

[0035] Fifthly, this application provides an electronic device including a processor coupled to a memory: the processor is configured to execute a computer program or instructions stored in the memory to cause the electronic device to perform the method as described in the first aspect and any possible implementation thereof, or to implement the method as described in the second aspect and any possible implementation thereof.

[0036] In a sixth aspect, this application provides a vehicle including units for implementing the method as described in the first aspect and any possible implementation of the first aspect, or for implementing the method as described in the second aspect and any possible implementation of the second aspect.

[0037] In a seventh aspect, this application provides a readable storage medium including a program or instructions that, when executed, are performed as described in the first aspect and any possible implementation thereof, or, when executed, are performed as described in the second aspect and any possible implementation thereof.

[0038] Eighthly, this application provides a computer program product that, when run on a computer, causes the computer to perform the method as described in the first aspect and any possible implementation thereof, or causes the computer to perform the method as described in the second aspect and any possible implementation thereof.

[0039] In a ninth aspect, embodiments of this application provide a communication system including a first sensor for implementing the method as described in the first aspect and any possible implementation of the first aspect, and a target node for implementing the method as described in the second aspect and any possible implementation of the second aspect.

[0040] In a tenth aspect, embodiments of this application provide a terminal device, including units for implementing the method as described in the first aspect and any possible design of the first aspect, or for implementing the method as described in the second aspect and any possible design of the second aspect. For example, the terminal device includes, but is not limited to: intelligent transportation equipment (such as automobiles, ships, drones, trains, freight trucks, etc.), intelligent manufacturing equipment (such as robots, industrial equipment, intelligent logistics, intelligent factories, etc.), and intelligent terminals (mobile phones, computers, tablets, PDAs, desktop computers, headphones, speakers, wearable devices, in-vehicle equipment, etc.).

[0041] Based on the implementations provided in the above aspects, the embodiments of this application can be further combined to provide more implementations.

[0042] The technical effects that can be achieved by any possible implementation of any of the second to tenth aspects above can be described with reference to the technical effects that can be achieved by any possible implementation of the first aspect above, and the repetitions will not be discussed. Attached Figure Description

[0043] Figure 1 The three-layer interface layer of ISO-23150 is shown;

[0044] Figure 2 The text interface types supported by the logical interface layer between a single sensor / a cluster of sensors and a fusion unit are shown.

[0045] Figure 3 A schematic diagram illustrating the application scenarios to which the embodiments of this application are applicable is shown;

[0046] Figure 4 This application illustrates a schematic diagram of the perception principle in the vehicle field according to an embodiment of the present application.

[0047] Figure 5 A flowchart illustrating the sensing method according to an embodiment of this application is shown;

[0048] Figure 6 The signal processing flow of the radar sensor and the different sensing data involved are shown.

[0049] Figure 7 A schematic diagram of a ROI-based reporting mode combination according to an embodiment of this application is shown;

[0050] Figure 8 The diagram shows the detection area under different scenarios according to embodiments of this application;

[0051] Figures 9A-9C The flowcharts of different examples of perception methods in embodiments of this application are shown.

[0052] Figure 10 A schematic diagram of a communication device according to an embodiment of this application is shown;

[0053] Figure 11 This paper shows another structural schematic diagram of a communication device according to an embodiment of the present application. Detailed Implementation

[0054] Currently, the perception results output by sensors or perception systems come in various forms and are not entirely uniform. Taking the vehicle field as an example, in low-level intelligent driving scenarios, for partially autonomous driving (AD) functions, a simple environmental perception model can be generated using only specific objects (such as vehicles, pedestrians, road markings, etc.). However, for advanced autonomous driving functions, it is necessary not only to merge the identified objects but also to include the specific attributes and characteristics of other sensors of these objects to generate a coherent model of the surrounding environment. This forms the basis of the dynamic surround model based on full-element information during vehicle autonomous driving. On this basis, the decision layer of the autonomous driving control system (such as the fusion unit) can observe the vehicle's spatiotemporal position in the driving environment, thereby making more accurate and reliable control decisions and plans.

[0055] To minimize the development work for sensor and sensing system suppliers and improve the reusability of development and verification work for different functions, the International Organization for Standardization (ISO) defines standardized logical interfaces for sensors and fusion units, denoted as ISO-23150. For example... Figure 1 As shown, ISO-23150 includes three interface layers, namely:

[0056] (1) The logical interface layer between the fusion unit and the AD function (e.g., represented as AD function 1, AD function 2 and AD function 3);

[0057] (2) Logical interface layer between a single sensor / a cluster of sensors and a fusion unit;

[0058] (3) The interface layer at the raw data level of the sensor's sensing element.

[0059] Among them, the text interface types supported by the logical interface layer between a single sensor / a certain type of sensor cluster and the fusion unit are as follows: Figure 2 As shown, it includes object-level (or target-level) interfaces, feature-level interfaces, and detection-level interfaces (such as radar point cloud-level interfaces).

[0060] At the object level, a single sensor or a cluster of sensors can use the following interface to provide the fusion unit with sensing data for different objects (or targets). This data can also be called object-level data or target-level data:

[0061] —The Potential Moving Object Interface (7.3) provides perception data for potential moving objects, including vehicles, people and other dynamic traffic participants;

[0062] —The Road Object Interface (7.4) is used to provide perception data for road objects, which include road surface, road markings, and road boundaries;

[0063] —The Static Objects Interface (7.5) is used to provide perception data for static objects, including general landmarks, traffic signs, and traffic lights.

[0064] At the feature level, a single sensor or a cluster of sensors can use the following interface to provide sensing data based on different features to the fusion unit. This data can also be referred to as feature-level data:

[0065] —Camera feature interface (8.3), used to provide perceptual data of camera features;

[0066] —Ultrasonic feature interface (8.4), used to provide sensing data of ultrasonic features;

[0067] At the detection level, a single sensor or a cluster of sensors of a certain type can use the following interface to provide sensing data based on different detection types to the fusion unit. This data can also be referred to as detection-level data:

[0068] —Radar detection interface (9.3), used to provide radar-based sensing data;

[0069] — LiDAR detection interface (9.4), used to provide sensing data based on LiDAR detection;

[0070] —Camera detection interface (9.5), used to provide perception data based on camera detection;

[0071] —Ultrasonic testing interface (9.6) for providing international and ultrasonic testing sensing data.

[0072] Optionally, a single sensor or a cluster of sensors can also provide supportive information to the fusion unit through the following interfaces:

[0073] —Sensor Performance Interface (10.3): Used to provide specific sensor attributes, such as current field of view (FOV) segment, object recognition rate, and reference target recognition rate;

[0074] — Sensor Health Information Interface (10.4): Used to provide specific attributes, such as general sensor status, calibration information, and sensor cluster definition information.

[0075] A fusion unit can combine perception data from multiple sensors (or clusters of multiple types of sensors) to improve the perception of targets or the environment. For example, in the automotive field, multi-sensor data fusion can compensate for the insufficient perception capabilities of a single sensor, providing richer environmental information for the vehicle's intelligent driving system. This facilitates more precise decision-making and control by the intelligent driving system, ensuring the safety and convenience of driving.

[0076] With the development of new energy vehicles and advanced driving assistance systems (ADAS), the demands for sensor data transmission have changed accordingly, and the data transmission interfaces defined by ISO-23150 are no longer sufficient. For example, in some vehicle designs, the network communication bandwidth and transmission latency of vehicles using Controller Area Network (CAN) / Flexible Data Rate (CAN-FD) cannot meet the requirements for enabling a single sensor / a cluster of sensors to simultaneously report richer perception information, such as simultaneously reporting target-level data and detection-level data. In other vehicle designs, while meeting the requirements for communication bandwidth and transmission latency, according to... Figure 2The single type of interface shown reports the perception results to the fusion unit, which cannot guarantee that the information can meet the requirements of the fusion unit for the data to be processed, nor can it guarantee that the performance of the vehicle's intelligent driving system is better.

[0077] To address the aforementioned problems, this application provides a sensing method and apparatus to enable sensors to provide more diverse sensing information to meet the demands of different scenarios, thereby improving processing efficiency and satisfying the needs of intelligent driving in various scenarios. The method and apparatus are based on the same technical concept. Since the principles by which the method and apparatus solve problems are similar, the implementation of the apparatus and method can be mutually referred to, and repeated details will not be repeated. Furthermore, in the various embodiments of this application, unless otherwise specified or logically conflicting, the terminology and / or descriptions between the various embodiments are consistent and can be mutually referenced. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0078] It should be noted that the perception scheme in this application embodiment can be applied to the Internet of Vehicles, such as vehicle-to-everything (V2X), long-term evolution-vehicle (LTE-V), and vehicle-to-vehicle (V2V). For example, it can be applied to vehicles with driving mobility functions, or other devices with driving mobility functions in vehicles. These other devices include, but are not limited to, on-board terminals, on-board controllers, on-board modules, on-board components, on-board chips, on-board units, on-board radar, or on-board cameras, and other sensors. Vehicles can implement the perception method provided in this application through these on-board terminals, on-board controllers, on-board modules, on-board components, on-board chips, on-board units, on-board radar, or on-board cameras. Of course, the control scheme in this application embodiment can also be used in other intelligent terminals with mobility control functions besides vehicles, or installed in other intelligent terminals with mobility control functions besides vehicles, or installed in components of such intelligent terminals. These intelligent terminals can be intelligent transportation equipment, smart home devices, robots, etc. Examples include, but are not limited to, smart terminals or controllers, chips, radar or cameras, and other sensors and components within smart terminals.

[0079] It should be noted that in the embodiments of this application, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can represent: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, and c can be single or multiple.

[0080] Furthermore, unless otherwise specified, the ordinal numbers such as "first" and "second" mentioned in the embodiments of this application are used to distinguish multiple objects and are not used to limit the priority or importance of multiple objects. For example, "first sensor" and "second sensor" are only used to distinguish different sensors, and do not indicate that the two sensors have different priorities or importance. In some embodiments, the method steps implemented by the first sensor and the second sensor can be interchanged.

[0081] For ease of understanding, the following description is provided in conjunction with the accompanying drawings and embodiments.

[0082] Figure 3 A schematic diagram of a communication system to which embodiments of this application apply is shown. This communication system may include a sensor 310 and a target node 320. The target node may also be referred to as a computing platform or computing node, and for example includes at least one processor 321, which can execute instructions 323 stored in a non-transitory computer-readable medium such as memory 322. In addition to instructions 323, memory 322 may also store data.

[0083] The product form or deployment method of the target node 320 can vary depending on the application scenario. For example, the target node 320 can be a standalone device, such as a roadside unit or a cloud server. Alternatively, the target node 320 can be a chip or component in a vehicle, or it can be a software module that can be deployed on relevant vehicle-mounted equipment, or it can be deployed on other types of terminal devices.

[0084] For example, in some embodiments, the target node can be implemented as a vehicle's computing platform, which can also be multiple computing devices controlling individual components or subsystems of the vehicle in a distributed manner. Processor 321 can be any conventional processor, such as a central processing unit (CPU). Alternatively, processor 321 can also include graphics processing units (GPUs), field-programmable gate arrays (FPGAs), system-on-chips (SoCs), application-specific integrated circuits (ASICs), or combinations thereof. Sensor 310 can be located on the same vehicle as the target node 320 or on a different vehicle. Data stored in memory 322 may include, for example, road maps, route information, vehicle position, direction, speed, and other such vehicle data, as well as other information. This information can be used by the vehicle and computing platform during operation in autonomous, semi-autonomous, and / or manual modes to assist the vehicle in achieving autonomous or intelligent driving.

[0085] In other embodiments, the target node 320 can be implemented as a sensor other than sensor 310, and possess computing capabilities. It can obtain more diverse perceptual information through information interaction with sensor 310 to compensate for its own limitations in perception. For example, sensor 310 can be a radar sensor, and target node 320 can be a camera with computing capabilities. When target node 320 is also implemented as a sensor, it can be installed on the same vehicle as sensor 310, or it can be installed on a different vehicle.

[0086] The sensing method in this embodiment can be implemented collaboratively by a target node 320 and a sensor 310. The sensor 310 can report at least one type of sensing data to the target node 320 through a data plane interface (or data transmission interface). The target node 320 can serve as... Figure 1 The fusion unit in the middle is used to receive at least one type of sensing data from the sensor 310, and can perform data fusion processing or decision control based on the received at least one type of sensing data, so that the sensor 310 can provide more diverse sensing information to meet the requirements of different scenarios, thereby improving processing efficiency and meeting the sensing needs under different scenarios.

[0087] like Figure 4As shown, the target node can be implemented as a computing node on the vehicle. This computing node can be an intelligent driving domain control unit (e.g., a mobile data center (MDC)), a vehicle control unit (VCU), or a vehicle domain controller (VDC) integrated into the vehicle's computing platform. This application does not limit this specific implementation. The sensors in the vehicle's sensing system can include, but are not limited to, cameras, LiDAR (light detection and ranging), millimeter-wave radar (RADAR), ultrasonic radar, infrared radar, or other sensors. Any of these sensors can be used as... Figure 3 The sensor 310 is used to acquire perception data of the vehicle's environment, constantly monitoring and collecting data about the vehicle's surroundings. When implementing the perception method of this embodiment, the sensor 310 can report at least one type of perception data to the vehicle's computing node through the data plane interface between the sensor and the vehicle's computing node. As an example, this at least one type of perception data may include at least one of the following: raw perception data; transmit / receive channel data; point cloud-level data; feature-level data; and target-level data. Different levels of perception data will be described below with reference to examples, and will not be elaborated upon here.

[0088] Accordingly, the vehicle's computing node can receive at least one type of perception data from sensors and perform perception processing based on this data. For example, the vehicle's computing node can perform data fusion processing on the received at least one type of perception data to obtain information about the surrounding environment of the vehicle's location. Alternatively, the vehicle's computing node can make decision-making and control decisions based on the received at least one type of perception data, generating control commands. Optionally, the vehicle's computing node can send control commands to the corresponding control devices through an in-vehicle communication network (e.g., a gateway), so that the relevant devices of the vehicle can assist in controlling the vehicle to achieve intelligent driving based on the received control commands. Meanwhile, the vehicle's sensing system may also include, but is not limited to, speed sensors, acceleration sensors, angular velocity sensors, roll angle sensors, steering wheel sensors, etc. During the process of the relevant devices of the vehicle assisting in controlling the vehicle to achieve intelligent driving, the vehicle's driving parameters can also be obtained through the corresponding sensors in the vehicle's sensing system, such as driving speed, wheel speed, longitudinal (lateral) acceleration, yaw rate, roll angle, steering wheel angle, heading angle, accelerator pedal opening information, brake pedal opening information, gear, driving mode, road mode, and battery state of charge (SOC). The vehicle's computing nodes can acquire various driving parameters of their respective vehicles and make vehicle control decisions based on these parameters.

[0089] In one optional implementation, the vehicle's computing node can also output alert information to peripheral devices such as touchscreens and speakers associated with the vehicle via a gateway, dynamically instructing the driver on the results of intelligent driving control decisions. Alternatively, the vehicle's computing node can also receive control information from the driver via (or through a gateway) peripheral devices such as touchscreens and microphones. This control information can be associated with the aforementioned alert information and can be used to assist the vehicle's computing node in making vehicle control decisions.

[0090] Figure 4 In this embodiment, the bidirectional arrows between different modules are only used to indicate that the corresponding modules can communicate with each other, and do not limit any communication method or information format. The target node can use different communication methods or information formats to communicate with different modules, and the target node can also have protocol conversion or format conversion functions, but this embodiment does not limit this. Figure 4 The other modules shown in the vehicle are merely examples; the dashed boxes indicate that the corresponding modules are optional and may not be included in the vehicle. Figure 4 The modules shown may also include, except for Figure 4 Other modules besides the ones shown, or will Figure 4Some modules can be replaced with other modules not shown, which will not be described in detail here. In some designs, the vehicle's sensing system can also be integrated into any one of MDC, VCU, or VDC, or VCU or VDC can also be integrated into MDC. The embodiments of this application do not limit the product form or integration method of different modules of the vehicle.

[0091] For ease of distinction, in the following text, a sensor or a cluster of sensors that provides at least one type of sensing data to a target node will be referred to as a first sensor, and a sensor or a cluster of sensors that can be implemented as a target node will be referred to as a second sensor. The first sensor can provide the second sensor with at least one type of sensing data that meets the requirements. When the first sensor implements the sensing method of the embodiments of this application, as follows: Figure 5 As shown, the following steps may be included:

[0092] S510: The first sensor acquires the first data.

[0093] In this embodiment of the application, the first data may include one or more of various sensing data. The first sensor can use the first data to provide more diverse sensing data to the target node to meet the requirements of different scenarios or to meet the indication requirements of different computing nodes.

[0094] As an example, the first data may include at least one of the following: raw sensing data; transmit / receive channel data; point cloud-level data; feature-level data; target-level data.

[0095] The raw sensing data refers to the raw data collected by the sensing element of the first sensor, such as analog-to-digital converter (ADC) data. The transmit / receive channel data is data related to the transmit / receive channels of the first sensor, and may include, for example, the distance between the transmit and receive antennas of the first sensor. The point cloud is the sampled points obtained after the first sensor detects an object. Point cloud data refers to a set of vectors in a three-dimensional coordinate system. Point cloud-level data is sensing data at the point cloud level (or granularity, dimension, etc.), and is an example of detection-level data. It is typically a single frame of point cloud data, and may include, for example, the orientation, distance, and radial velocity information of the target object. Feature-level data is sensing data at the feature level, and may include data from camera features, ultrasonic features, radar features, etc. The first sensor can use target detection technology to identify target objects in the environment, or it can use target tracking technology to provide the target's motion trajectory. Target-level data may include, for example, multiple frames of point cloud data of the target object, including orientation, distance, absolute velocity, and size.

[0096] In one optional implementation, the first sensor may support different reporting modes, and the content of the first data may be related to the reporting mode. The reporting mode refers to the mode used by the first sensor to report sensing data to the target node. Specifically, it may refer to the data type mode of one or more reported sensing data, the encapsulation / combination mode of one or more reported sensing data, or the text interface type mode of one or more reported sensing data. This application embodiment does not limit this.

[0097] As an example, the reporting modes supported by the first sensor may include at least one of the following: raw sensing data reporting mode; transmit / receive channel data reporting mode; point cloud-level data reporting mode; feature-level data reporting mode; and target-level data reporting mode. Specifically, the raw sensing data reporting mode can be a mode for reporting raw sensing data. The transmit / receive channel data reporting mode can be a mode for reporting transmit / receive channel data. The point cloud-level data reporting mode can be a mode for reporting point cloud-level data. The feature-level data reporting mode can be a mode for reporting feature-level data. The target-level reporting mode can be a mode for reporting target-level data. The first sensor can use any of the above reporting modes to report the corresponding sensing data to the target node.

[0098] In an optional implementation, the reporting mode of the first data can also be a combination of at least one of the above reporting modes. When implementing S510, the first sensor can acquire data corresponding to at least one reporting mode as needed, as the content of the first data, so as to report more diverse perception information to the target node to meet the requirements of different scenarios.

[0099] It should be understood that the reporting modes and corresponding data classification methods described here are merely illustrative and not limiting. In specific implementations, the reporting modes supported by the first sensor may vary depending on its own sensing and computing capabilities (or data processing capabilities). For example, in some embodiments, the reporting modes supported by the first sensor may also include a detection-level data reporting mode, where detection-level data may include, for example, point cloud-level data. In other embodiments, the reporting modes supported by the first sensor may also include an intermediate data reporting mode, where intermediate data may be data processed by the first sensor on the raw sensing data but not processed to its final form, where the final form of the data may be, for example, target-level data.

[0100] Taking the first sensor as an example, such as a radar sensor Figure 6 As shown, the radar signal processing flow can include:

[0101] ADC data → One-dimensional Fast Fourier Transform (FFT) processing (represented as 1DFFT or 1 st FFT → Two-dimensional FFT processing (represented as 2D FFT or 2 nd FFT), constant false alarm rate detector (CFAR) processing → target detection (range measurement, velocity measurement, angle measurement) → target tracking.

[0102] The ADC data can be an example of raw sensing data. A series of processes performed by the radar sensor on the raw sensing data can yield different data, such as 1. st The output of the FFT (Distance FFT) can provide the beat frequency, amplitude, and phase of each target object. Based on 1 st The FFT output is 1 st FFT selection yields 1D FFT data, including range bins generated from multiple samples. This 1D FFT data serves as an example of detection-level data. Based on 2 nd The FFT output, after undergoing CFAR processing, can monitor the noise of each or each range block, compare the signal with the local noise level, and automatically adjust the radar sensitivity to maintain a constant false alarm rate. This process yields CFAR data from the radar sensor, serving as an example of intermediate data. Following CFAR processing, target detection processing such as ranging, velocity measurement, and angle measurement is performed to obtain direction of arrival (DoA) data, or detection-level data such as distance and angle between the first sensor and the target object. This detection-level data may include, for example, point cloud data of the target object. Further target tracking processing yields target-level data, such as multi-frame point cloud data including one or more specific targets. The data processed by CFAR, target detection, and target tracking can also be used for... st FFT selection yields 1D FFT data.

[0103] The raw sensing data or data obtained after different processing steps involved in the above processing flow can all be used as sensing data to be reported. For example, a radar sensor can report raw sensing data to the target node based on the raw sensing data reporting mode. Alternatively, a radar sensor can report target-level data to the target node based on the target-level data reporting mode. Alternatively, a radar sensor can report transmit / receive channel data to the target node based on the transmit / receive channel data reporting mode. Alternatively, a radar sensor can report detection-level data to the target node based on the detection-level data reporting mode, where the detection-level data includes point cloud-level data or 1D FFT data. Alternatively, a radar sensor can report point cloud-level data to the target node based on the point cloud-level data reporting mode. Alternatively, a radar sensor can report feature-level data to the target node based on the feature-level data reporting mode. Alternatively, a radar sensor can report CFAR data to the target node based on the intermediate data reporting mode, etc.

[0104] Because the information carried by the different sensing data reported by the first sensor to the target node under different reporting modes varies, it can affect the corresponding data fusion processing results or decision control results. For example, if the first sensor reports target-level data to the target node, the target node can easily obtain information about target objects in the surrounding environment. Or, for example, point clouds contain more information; a single target object may have multiple frames of point cloud data. If the first sensor reports the point cloud-level data of the target object to the target node, the target node can expand and obtain more target information based on the point cloud-level data. Or, for example, transmit / receive channel data can be used for micro-Doppler extraction. If the first sensor reports transmit / receive channel data to the target node, the target node can easily obtain more point cloud data. Or, for example, if the first sensor reports raw sensing data to the target node, the latency of the target node acquiring sensing data can be reduced. By leveraging the target node's own powerful computing capabilities to process the raw sensing data, the latency caused by data processing can be reduced.

[0105] In order to address the changing data transmission requirements of sensors due to the development of ADAS, in this embodiment of the application, the first sensor supports different reporting modes, and the content of the first data is related to the reporting mode. When the first sensor implements S510, it can acquire data corresponding to at least one reporting mode as needed, and use it as the content of the first data.

[0106] For example, the first data may include at least one of the following sensing data: raw sensing data; transmit / receive channel data; point cloud-level data; feature-level data; target-level data. In other examples, the content of the first data may also include sensing data corresponding to one or more other reporting modes; this application embodiment does not limit the content of the first data. The first sensor is used as an example. Figure 6Taking a radar sensor as an example, the first data may include Figure 6 The at least one type of perception data shown, such as the first data, may include target-level data and detection-level data. The detection-level data may include 1D FFT data or point cloud-level data of one or more target objects. Alternatively, the first data may include target-level data, point cloud-level data, and CFAR data. Alternatively, the first data may include point cloud-level data and raw perception data. Alternatively, the first data may include point cloud-level data, raw perception data, and CFAR data.

[0107] Similarly, when the first sensor is implemented as a single sensor or a single-type sensor cluster of other types, the first sensor can also implement S510 according to its own capabilities to obtain the sensing data corresponding to at least one reporting mode as the content of the first data, so as to report diverse sensing information to the target node to meet the requirements of different scenarios. The embodiments of this application do not limit the type of the first sensor, the reporting mode supported by the first sensor, or the content of the first data.

[0108] S520: The first sensor sends first data to the target node. Accordingly, the target node can receive the first data from the first sensor and perform sensing processing based on the first data.

[0109] For example Figure 6 As shown, when the target node is implemented as a computing node of the vehicle, the radar sensor, acting as the first sensor, can send first data to the computing node through the data plane interface between them. The computing node can then perform data fusion processing or decision control based on the received first data. Similarly, when the target node is implemented as a second sensor, the first sensor, implementing S520, can send first data to the second sensor through the data plane interface between them. The second sensor can then obtain more diverse target information based on the received first data to compensate for its own perception capabilities.

[0110] Therefore, through the above perception method, the first sensor, based on its supported different reporting modes, reports at least one type of perception data to the target node to provide more diverse perception information to meet the needs of different scenarios. This can satisfy the demand for a single sensor / a cluster of sensors to report richer perception information simultaneously in some scenarios, and also ensure that at least one type of perception data reported in different scenarios meets the target node's requirements for the data to be processed, thereby improving processing efficiency and meeting the needs of intelligent driving in different scenarios.

[0111] In this embodiment of the application, when implementing S510, the first sensor can acquire the content of the first data in different ways.

[0112] For example, the first sensor can use latency as an evaluation metric to decide how to select the content of the first data. For instance, based on latency, different types of sensed data can meet the following priority levels:

[0113] The raw sensing data has a higher priority than the transmit / receive channel data;

[0114] The priority of transmit / receive channel data is higher than that of the point cloud-level data;

[0115] Point cloud-level data has a higher priority than feature-level data and target-level data.

[0116] If the target node operates in a scenario with high latency requirements (i.e., low latency requirements), the first sensor can report more unprocessed sensing data to the target node, such as raw sensing data and / or transmit / receive channel data, to reduce the latency of data acquisition for the target node. The target node can then process this raw sensing data using its own computing power, further reducing latency caused by data processing. If the target node operates in a scenario with low latency requirements (i.e., high latency requirements), the first sensor can report more data that has undergone one or more levels of processing, such as at least one of feature-level data, point cloud-level data, and target-level data. Furthermore, the target node can directly perform data fusion processing or decision control based on the received at least one type of data. Similarly, for other types of sensors, depending on their sensing and computing capabilities, at least one type of sensing data to be reported can be acquired as the content of the first data and sent to the target node.

[0117] Alternatively, for example, the first sensor can report one or more types of sensing data according to different regions of interest (ROIs) within its detection area. For instance, the first data may include sensing data for different ROIs within the first sensor's detection area, with different reporting modes corresponding to different ROIs. The first sensor can determine the different ROIs within its detection area based on its field of view, detection distance range, or AD function. Alternatively, the first sensor can receive third indication information from the target node, indicating the different ROIs within its detection area. The reporting modes corresponding to different ROIs can be fixed or adaptively adjusted, and the first sensor can obtain the content of the first data based on the reporting modes corresponding to different ROIs.

[0118] Taking the first sensor being located at the front of the vehicle as an example, such as Figure 7As shown, the detection area of ​​the first sensor can be a sector-shaped area centered at the location of the first sensor (denoted as O), with the X-axis parallel to the vehicle's lateral position and the Y-axis parallel to the vehicle's longitudinal position and pointing towards the vehicle's forward direction. The detection area of ​​the first sensor can be divided into multiple Regions of Interest (ROIs) relative to the center O, denoted as Region ①, Region ②, Region ③, and Region ④. Different symbols are used to represent different types of sensing data output; for example, a parallelogram represents raw sensing data output, a triangle represents transmit / receive channel data output, a hollow circle represents point cloud-level data output, and a rectangle represents target-level data output. As an example, each ROI can correspond to at least one reporting mode, and different ROIs can correspond to different reporting modes. For example, Region ① corresponds to the transmit / receive channel data reporting mode, Region ② corresponds to the point cloud-level data reporting mode, and Regions ③ and ④ correspond to the target-level data reporting mode. Accordingly, the first data obtained by the first sensor during S510 can include: transmit / receive channel data in Region ①, point cloud-level data in Region ②, and target-level data in Regions ③ and ④. Alternatively, within the detection area of ​​the first sensor, the sensing data for different ROIs can be a combination of distance and azimuth information, or a combination of horizontal and vertical position information.

[0119] In one optional implementation, the reporting mode corresponding to different ROIs can also be related to the priority of the sensing data. Different types of sensing data can meet the following priority levels: the priority of raw sensing data is higher than that of transmit / receive channel data; the priority of transmit / receive channel data is higher than that of point cloud-level data; and the priority of point cloud-level data is higher than that of feature-level data and target-level data. Based on the correlation between different ROIs and the vehicle's intelligent driving functions, the reporting mode corresponding to different ROIs can be determined based on priority. For example, region ① corresponds to the transmit / receive channel data reporting mode and the raw sensing data reporting mode (only one reporting mode corresponding to region ① is shown as an example in the figure), region ② corresponds to the point cloud-level data reporting mode, and regions ③ and ④ correspond to the target-level data reporting mode. Accordingly, the first data obtained by the first sensor when implementing S510 may include: transmit / receive channel data and raw sensing data in region ①, point cloud-level data in region ②, and target-level data in regions ③ and ④. In this way, for ROIs close to the first sensor (e.g., region ①), the first sensor reports data from the transmit / receive channels to help the target node obtain more point cloud data. For ROIs far from the first sensor (e.g., regions ③ and ④), the first sensor reports target-level data to help the target node directly obtain information about target objects in the surrounding environment.

[0120] It should be understood that this is merely an example of at least one reporting mode corresponding to different ROIs and the sensing data corresponding to the at least one reporting mode, and not a limitation. In other cases, the ROI division of the first sensor may be different, and the sensing data corresponding to different ROIs may be replaced with other types of sensing data.

[0121] For example Figure 8 As shown, the first sensor is installed on the vehicle. Depending on the specific road conditions, the vehicle is located in lane 1, which is not a one-way lane. There are lanes 2 and / or 3 adjacent to lane 1 and in the same direction on the road where lane 1 is located. According to the driving intention of the vehicle driver or the driving intention of the autonomous driving system, when the vehicle moves forward in the driving direction, the detection area of ​​the first sensor can be a first area in front of the vehicle, including the area of ​​lane 1 in front of the vehicle when the vehicle moves forward in the driving direction, as well as the areas of lane 2 and / or lane 3.

[0122] like Figure 8 As shown in the rectangular filled box, the range of the first region differs in cases (1), (2), or (3). In case (1), the first region includes the area of ​​lane 1 within a predetermined distance in front of the vehicle when it travels in the direction of travel, and the area of ​​lane 2. The width of the first region can be the same as the sum of the widths of lane 1 and lane 2 (a preset width error is allowed). The length of the first region can be preset, and can be a numerical value (e.g., 10 meters) or a length range (e.g., 5 meters to 15 meters). In case (2), the first region can include the area of ​​lane 1 within a predetermined distance in front of the vehicle when it travels in the direction of travel, and the area of ​​lane 3. The width of the first region can be the same as the sum of the widths of lane 1 and lane 3 (a preset width error is allowed). The length of the first region can be preset, and can be a numerical value (e.g., 10 meters) or a length range (e.g., 5 meters to 15 meters). In scenario (3), the first area may also include the area of ​​lane 1, the area of ​​lane 2 and the area of ​​lane 3 within a predetermined distance in front of the vehicle when the vehicle is traveling in the direction of travel. The width of the first area may be the same as the sum of the widths of lane 1, lane 2 and lane 3 (a preset width error is allowed). The length of the first area may be preset, may be a numerical value (e.g. 10 meters) or may be a length range (e.g. 5 meters to 15 meters).

[0123] Figure 8The three scenarios described are merely examples of the shape, length, or width of the first region, not limitations. In practical applications, adjustments can be made based on actual needs or changes in road conditions, which will not be elaborated upon here. In some designs, vehicles may also change lanes. During lane changes, the detection area of ​​the first sensor and different ROIs can adaptively change, which will not be elaborated upon here. Alternatively, in some designs, the road where the vehicle is located may have intersections. In different intersection scenarios, the detection area of ​​the first sensor and different ROIs can adaptively change, which will not be elaborated upon here.

[0124] exist Figure 8 In the different situations shown, with Figure 7 Similarly, the fan-shaped detection area shown can also be divided into different ROIs within the first area, such as the ROI in front of the vehicle in the same lane, or the ROI in front of the vehicle in the adjacent lane; or, at a finer granularity, the detection area in front of the vehicle in the same lane can be further divided according to its distance relative to the first sensor into ROIs close to the first sensor (e.g., within 10 meters), ROIs relatively close to the first sensor (e.g., within 10-50 meters), and ROIs far from the first sensor (e.g., beyond 50 meters). Different ROIs can correspond to different reporting modes. When implementing S510, the first sensor can acquire the sensing data corresponding to different ROIs as the content of the first data. The detailed implementation of this method is combined with the above. Figure 7 The method shown is similar; for details, please refer to the previous text. Figure 7 The relevant descriptions will not be repeated here.

[0125] In one embodiment of this application, the reporting mode of the first data can be related to the scene in which the first sensor is located. The scene in which the first sensor is located can be differentiated based on time delay. When implementing S510, the first sensor can adaptively obtain the content of the first data according to its current scene, and report the obtained first data to the target node in S520. In another example, before implementing S510, the target node can send indication information to the first sensor. This indication information can indicate the target node's customized requirements for the sensing data. When implementing S510, the first sensor can obtain the content of the first data according to the received indication information, and report the obtained first data to the target node in S520. For ease of understanding, the following uses the example of the first sensor being installed on a vehicle as an example, combined with different examples for detailed description.

[0126] Example 1: such as Figure 9A As shown in the flowchart, the first sensor can acquire the content of the first data through internal adaptive adjustment and report it to the target node.

[0127] In Example 1, the first sensor can be pre-configured with a scene-based reporting strategy. The first sensor can determine the reporting strategy and reporting mode based on the scene in which the vehicle is located, and obtain the content of the first data. The scene-based reporting strategy can be pre-stored in a storage medium accessible to the first sensor, or the vehicle's computing node can pre-obtain the scene-based reporting strategy from a cloud server and store it in a storage medium accessible to the first sensor. This embodiment does not limit the pre-configuration method of these strategies.

[0128] In practical implementation, the scenario in which the vehicle is located can be related to the external environment of its location. This external environment can include static environments, such as obstacles, surrounding landscapes, traffic facilities, and roads. Alternatively, it can include dynamic environments, such as dynamic indicator facilities and communication environment information. Dynamic indicator facilities can include traffic lights and variable message signs, while communication environment information can include signal strength, electromagnetic interference, and signal delay information. External information can also include traffic participants, such as animals, pedestrians, and other vehicles, including both motorized and non-motorized vehicles. Furthermore, the external environment can include meteorological conditions, such as ambient temperature, lighting conditions, and weather conditions. Under different external environmental factors, the vehicle's computing nodes require different perceptual data when making intelligent driving control decisions. The first sensor can acquire and report the first data using different reporting modes in different scenarios to provide richer perceptual information to the vehicle's computing nodes to meet fusion requirements, thereby improving perception accuracy and ensuring that the vehicle's computing nodes can make precise intelligent driving decisions and controls. This ensures the accuracy of decision control and optimizes the overall system performance, ultimately guaranteeing safe driving.

[0129] Vehicle service providers can collect the demand for perception data from different environmental elements in advance, pre-set multiple scenarios and reporting strategies corresponding to each scenario. The reporting strategy can include a combination of at least one reporting mode. The first sensor can adaptively adjust the current scenario based on the perceived environmental information and obtain the content of the first data based on the reporting strategy corresponding to the scenario, including the perception data corresponding to at least one reporting mode.

[0130] As an example, the scenario in which the first sensor is located can include any of the following scenarios based on the type of road the vehicle is on: urban road scenario, highway scenario; or, the scenario in which the first sensor is located can include any of the following scenarios based on the weather conditions at the vehicle's location: good visibility scenario, moderate visibility scenario, poor visibility scenario; or, the scenario in which the first sensor is located can include any of the following scenarios based on the road conditions of the road segment the vehicle is on: severely congested road segment scenario, moderately congested road segment scenario, lightly congested road segment scenario, unobstructed road segment scenario. Different scenarios can be configured with fixed reporting mode combinations by default, such as combinations of at least one of the following reporting modes: raw perception data reporting mode; transmit / receive channel data reporting mode; point cloud level data reporting mode; feature level data reporting mode; target level data reporting mode. The first sensor obtains environmental information directly or indirectly to identify one or more scenarios in which the vehicle is located, and quickly and adaptively determines the corresponding reporting mode combination. Optionally, the first data can also include information on different ROIs, such as a combination of distance and azimuth information, or a combination of lateral and longitudinal position information.

[0131] In one optional implementation, the scenario in which the first sensor is located may also be related to the vehicle's driving parameters. These driving parameters may include, but are not limited to, vehicle speed, wheel speed, longitudinal (lateral) acceleration, yaw rate, roll angle, steering wheel angle, heading angle, accelerator pedal opening information, brake pedal opening information, gear position, driving mode, road mode, and battery state of charge (SOC). The vehicle's driving parameters can be acquired by at least one of the following sensors in the vehicle's sensing system: speed sensor, acceleration sensor, angular velocity sensor, roll angle sensor, steering wheel sensor, and other sensors. The first sensor can directly or indirectly obtain various driving parameters of the vehicle and determine the scenario in which the vehicle is located, so as to adaptively adjust the reporting mode combination to obtain at least one type of perception data that meets the scenario requirements as the content of the first data.

[0132] As an example, the scenario in which the first sensor is located can include any of the following scenarios based on the vehicle's driving speed: low-speed driving scenario, medium-speed driving scenario, and high-speed driving scenario. The first sensor can utilize acquired environmental information, driving parameters, etc., to identify one or more scenarios in which the vehicle is located, and quickly and adaptively determine the corresponding reporting mode combination. For example, in any of the above-mentioned low-speed driving scenario, urban road scenario, good visibility scenario, severely congested road segment scenario, or moderately congested road segment scenario, the content of the first data can include target-level data of different ROIs of the first sensor. In any of the following scenarios: medium-speed driving scenario, high-speed driving scenario, highway scenario, moderate visibility scenario, poor visibility scenario, lightly congested road segment scenario, or unobstructed road segment scenario, the content of the first data can include at least one of the following: raw sensing data of the first ROI, transmit / receive channel data, or point cloud-level data, and target-level data of the second ROI, wherein the detection distance corresponding to the first ROI is less than the detection distance corresponding to the second ROI.

[0133] In practical implementation, the vehicle's scenario can be comprehensively considered, taking into account factors such as the vehicle's driving parameters and external environmental elements. For example, in a scenario where the vehicle is traveling at low speed and the road is congested, the requirement for response latency is lower than in a high-speed driving scenario. The reporting mode combination associated with this scenario can include a target-level data reporting mode. In this scenario, the first sensor can acquire target-level data for different ROIs and report it to the target node as the content of the first data.

[0134] For example, in a scenario where the vehicle is traveling at high speed and braking, the response time to acceleration and deceleration of targets ahead is critical. The associated reporting modes in this scenario could include: for dynamic targets directly ahead in the main lane (e.g., the vehicle's lane), a target-level data reporting mode or a transmit / receive channel data reporting mode is used; for other ROIs, a target-level data reporting mode is used. In this scenario, the first sensor can acquire target-level data or transmit / receive channel data of dynamic targets directly ahead in the main lane, as well as target-level data of other ROIs, and report this as the content of the first data to the target node.

[0135] Alternatively, under specific weather conditions, such as rain or fog, the sensing capabilities of other sensors are reduced. The associated reporting modes for this scenario could include: a target-level data reporting mode, a transmit / receive channel data reporting mode, or a point cloud-level data reporting mode for the area directly ahead of the main lane (e.g., the lane the vehicle is in); and a target-level data reporting mode for other ROIs. In this scenario, the first sensor can acquire target-level data, transmit / receive channel data, or point cloud-level data for the ROI directly ahead of the main lane, as well as target-level data for other ROIs, and report this as the first data to the target node.

[0136] In one alternative implementation, the first sensor can also sense the computing power of the target node at the upper layer. When the computing power of the target node decreases, the first sensor can also adjust the content of the first data, for example, by adopting a target-level data reporting mode to report target-level data to the target node. By using the computing power of the first sensor itself, the computing pressure of the target node can be shared, thereby improving the efficiency of control decision-making.

[0137] Therefore, based on the different needs for perception data in different scenarios, in Example 1, the first sensor can adaptively adjust the combination of reporting modes according to the scenario to provide the target node with at least one type of perception data that is more diverse and meets the needs of intelligent driving control decision-making, thereby improving the perception accuracy of the first sensor, ensuring that the target node makes accurate intelligent driving decision-making and control, ensuring the accuracy of decision-making and control, and making the performance of the intelligent driving system as good as possible.

[0138] Example 2: such as Figure 9B As shown in the flowchart, the target node can be implemented as the vehicle's computing node. The first sensor can receive customized requirements from the vehicle's computing node, and obtain the content of the first data through an adaptive response and report it to the computing node.

[0139] like Figure 4 As shown, the target node can be implemented as a vehicle computing node, and a control plane interface can also be included between the vehicle computing node and the sensors. In Example 2, as... Figure 9B As shown, the target node (e.g., the vehicle's computing node) can send first indication information to the first sensor through this control plane interface to indicate the computing node's customized requirements for sensing data. For example, the first indication information may include at least one of the following: the vehicle's speed; the speeds of other vehicles near the vehicle; the geographical type of the detection area; the meteorological environment information corresponding to the detection area; and the load information of the vehicle's computing node. Correspondingly, the first sensor can receive the first indication information from the target node, and the first sensor can determine the scene in which it is located based on the first indication information. When the first sensor is installed on the vehicle, it can determine the scene in which the vehicle is located based on the first indication information. The reporting mode of the first data includes a combination of at least one of the following reporting modes: raw sensing data reporting mode; transmit / receive channel data reporting mode; point cloud-level data reporting mode; feature-level data reporting mode; and target-level data reporting mode. The specific implementation is similar to the description in conjunction with Example 1 above; please refer to the relevant description in conjunction with Example 1 above, which will not be repeated here.

[0140] In one optional implementation, the vehicle's computing node can further send second indication information to the first sensor. This second indication information can indicate the reporting mode of the first sensor or the scene in which the first sensor is located. The second indication information indicating the reporting mode of the first sensor can, for example, indicate different Regions of Interest (ROIs) within the detection area of ​​the first sensor and the corresponding reporting modes for each ROI, or indicate different combinations of reporting modes and the ROIs within each mode. Accordingly, the first sensor can receive the second indication information from the vehicle's computing node and obtain the content of the first data based on this second indication information. Similarly, the reporting mode of the first data can include a combination of at least one of the following reporting modes: raw perception data reporting mode; transmit / receive channel data reporting mode; point cloud-level data reporting mode; feature-level data reporting mode; and target-level data reporting mode.

[0141] In practical implementation, the vehicle's computing node can send a first instruction to the first sensor based on driving conditions. For example, when a vehicle ahead is detected moving, a second instruction can be sent to the first sensor to customize the following reporting mode combinations: a point cloud-level data reporting mode or a transmit / receive channel data reporting mode for the area ahead, and a target-level data reporting mode for other ROIs. Accordingly, the first data obtained by the first sensor includes point cloud-level data or transmit / receive channel data for the area ahead, as well as target-level data for other ROIs. After receiving the first data reported by the first sensor, the vehicle's computing node can obtain more information about the extended target based on the point cloud-level data of the area ahead, or extract micro-Doppler information based on the transmit / receive channel data to obtain more point cloud data. The target-level data corresponding to other ROIs can be used directly.

[0142] Alternatively, for example, when the vehicle's computing node identifies a deficiency / shortcoming of the second sensor, such as glare from a lidar sensor or difficulties with long-range detection, it can send a second instruction to the first sensor to customize the data reporting mode for the transmit / receive channel. This allows the first sensor to report transmit / receive channel data to the computing node along with the first data. Or, for example, for uncommon vehicle types (referred to as non-standard vehicles), if the vehicle's computing node identifies that the second sensor (e.g., a camera) cannot accurately perceive the target, it can also send a second instruction to the first sensor (e.g., a radar sensor) to customize the point cloud-level data reporting mode. This allows the first sensor to report point cloud-level data of one or more ROIs to the computing node along with the first data, thereby improving perception accuracy.

[0143] In optional implementations, the first data may also include information on different ROIs, such as a combination of distance and azimuth information, or a combination of horizontal and vertical positions, which will not be elaborated here.

[0144] Therefore, based on the different needs of the target node for perception data, in Example 2, the target node can send a first instruction message or a second instruction message to the first sensor to customize the target node's different needs for perception data. This allows the first sensor to provide the target node with at least one more diverse perception data that meets the needs of intelligent driving control decisions based on the received instruction message, thereby improving the perception accuracy of the first sensor and ensuring that the target node can make accurate intelligent driving decisions and controls, ensuring the accuracy of decision control, and making the performance of the intelligent driving system as good as possible.

[0145] Example 3: such as Figure 9C As shown in the flowchart, the target node can be implemented as a second sensor. The first sensor can receive customized requirements from the second sensor and obtain the content of the first data through adaptive blind spot compensation and report it to the second sensor.

[0146] In this embodiment, different sensors can also support a lateral communication interface. The second sensor can send customized requirements to the first sensor through the lateral communication interface. The first sensor can report the required first data to the second sensor according to the customized requirements of the second sensor, so as to fill the blind spots in the sensing range or sensing capability of the second sensor.

[0147] In Example 3, such as Figure 9C As shown, the second sensor can send second indication information to the first sensor via a lateral communication interface (an example of a control plane interface). This second indication information can indicate the reporting mode of the first sensor or the scene in which the first sensor is located. The second indication information indicating the reporting mode of the first sensor can, for example, indicate different Regions of Interest (ROIs) within the detection area of ​​the first sensor and the corresponding reporting modes for each ROI, or indicate different combinations of reporting modes and the ROIs within each mode. Accordingly, the first sensor can receive the second indication information from the second sensor and obtain the content of the first data based on this second indication information. Similarly, the reporting mode of the first data can include a combination of at least one of the following reporting modes: raw sensing data reporting mode; transmit / receive channel data reporting mode; point cloud-level data reporting mode; feature-level data reporting mode; and target-level data reporting mode.

[0148] In practical implementation, the second sensor, for example, when detecting a false alarm due to dirt or occlusion, sends a second instruction to the first sensor to customize a point cloud-level data reporting mode or a target-level data reporting mode for the false alarm area. The first sensor can then report the point cloud-level data or target-level data corresponding to the false alarm area to the second sensor, carrying it in the first data. Alternatively, during calibration, the second sensor can send a second instruction to the first sensor to customize a point cloud-level data reporting mode. Accordingly, the first sensor can report the point cloud-level data of each ROI to the second sensor, carrying it in the first data, to assist the second sensor in rapid calibration.

[0149] In optional implementations, the first data may also include information on different ROIs, such as a combination of distance and azimuth information, or a combination of horizontal and vertical positions, which will not be elaborated here.

[0150] Therefore, based on the different data requirements of the second sensor, in Example 3, the second sensor can send a second instruction to the first sensor to customize its different data requirements. This allows the first sensor to provide richer data to the second sensor based on the received second instruction, compensating for blind spots in the second sensor's perception capabilities and improving its perception accuracy. Optionally, the second sensor can also have computing capabilities, enabling it to perform data fusion processing or decision control based on the data provided by the first sensor.

[0151] This application also provides a communication device for executing the method performed by the first sensor or target node in the above method embodiments. The relevant features can be found in the above method embodiments, and will not be repeated here.

[0152] like Figure 10 As shown, the communication device 1000 may include an acquisition unit 1001 and a transceiver unit 1002. Optionally, the communication device 1000 may include a processing unit 1003.

[0153] In one example, acquisition unit 1001 is used to acquire first data, wherein the first sensor supports different reporting modes, and the content of the first data is related to the reporting mode; transceiver unit 1002 is used to send the first data to the target node; wherein the first data includes at least one of the following sensing data: raw sensing data; transceiver channel data; point cloud level data; feature level data; target level data. For specific implementation methods, please refer to the method steps implemented by the first sensor in the above method embodiments, which will not be repeated here.

[0154] In another example, transceiver unit 1002 is used to receive first data from a first sensor, wherein the first sensor supports different reporting modes, and the content of the first data is related to the reporting mode; processing unit 1003 is used to perform perception processing based on the first data; wherein the first data includes at least one of the following perception data: raw perception data; transceiver channel data; point cloud-level data; feature-level data; target-level data. For specific implementation details, please refer to the method steps implemented by the target node in the above method embodiments, which will not be repeated here.

[0155] It should be understood that the division of units in the above device is only a logical functional division. In actual implementation, they can be fully or partially integrated into a single physical entity, or they can be physically separated. Furthermore, the units in the device can be implemented by a processor calling software; for example, the device includes a processor connected to a memory containing instructions. The processor calls the instructions stored in the memory to implement any of the above methods or to implement the functions of each unit in the device. The processor can be, for example, a general-purpose processor, such as a Central Processing Unit (CPU) or a microprocessor, and the memory can be internal or external to the device. Alternatively, the units in the device can be implemented as hardware circuits. The functionality of some or all units can be achieved through the design of these hardware circuits, which can be understood as one or more processors. For example, in one implementation, the hardware circuit is an application-specific integrated circuit (ASIC). The functionality of some or all of the above units is achieved through the design of the logical relationships between the components within the circuit. In another implementation, the hardware circuit can be implemented using a programmable logic device (PLD). Taking a field-programmable gate array (FPGA) as an example, it can include a large number of logic gates. The connection relationships between the logic gates are configured through a configuration file, thereby achieving the functionality of some or all of the above units. All units of the above device can be implemented entirely through processor-invoked software, entirely through hardware circuits, or partially through processor-invoked software with the remaining parts implemented through hardware circuits.

[0156] As can be seen, each unit in the above device can be one or more processors (or processing circuits) configured to implement the above methods, such as: CPU, GPU, NPU, TPU, DPU, microprocessor, DSP, ASIC, FPGA, or a combination of at least two of these processor forms.

[0157] Furthermore, the units in the above devices can be integrated in whole or in part, or they can be implemented independently. In one implementation, these units are integrated together as a system-on-a-chip (SOC). The SOC may include at least one processor for implementing any of the above methods or implementing the functions of the units in the device. The at least one processor may be of different types, such as CPU and FPGA, CPU and artificial intelligence processor, CPU and GPU, etc.

[0158] In a simplified embodiment, those skilled in the art will realize that the communication devices described in the above embodiments can all be employed. Figure 11 As shown in the figure.

[0159] like Figure 11 The illustrated device 1100 includes at least one processor 1110 and a communication interface 1130. In an alternative design, a memory 1120 may also be included.

[0160] The specific connection medium between the processor 1110 and the memory 1120 is not limited in the embodiments of this application.

[0161] In such Figure 11 In the device, when the processor 1110 communicates with other devices, it can transmit data through the communication interface 1130.

[0162] When the communication device adopts Figure 11 When in the form shown, Figure 11 The processor 1110 can call computer execution instructions stored in the memory 1120, so that the device 1100 can execute any of the above method embodiments.

[0163] This application also relates to a chip system including a processor for calling a computer program or computer instructions stored in a memory to cause the processor to execute the methods of any of the above embodiments.

[0164] In one possible implementation, the processor can be coupled to the memory via an interface.

[0165] In one possible implementation, the chip system may also directly include a memory in which computer programs or computer instructions are stored.

[0166] For example, the memory can be volatile memory or non-volatile memory, or may include both. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which serves as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (SLDRAM), and direct rambus RAM (DR RAM).

[0167] This application also relates to a processor for calling a computer program or computer instructions stored in a memory to cause the processor to execute the methods described in any of the above embodiments.

[0168] For example, in the embodiments of this application, the processor is an integrated circuit chip with signal processing capabilities. For instance, the processor can be an FPGA, a general-purpose processor, a DSP, an ASIC, or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, a system-on-chip (SoC), a CPU, a network processor (NP), a microcontroller unit (MCU), a PLD, or other integrated chips, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in the embodiments of this application. A general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. This storage medium is located in memory; the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above methods.

[0169] It should be understood that embodiments of this application may be provided as methods, systems, or computer program products.

[0170] In one possible implementation, embodiments of this application provide a computer-readable storage medium storing program code that, when executed on a computer, causes the computer to perform the method embodiments described above.

[0171] In one possible implementation, this application provides a computer program product that, when run on a computer, causes the computer to execute the above-described method embodiments.

[0172] Therefore, this application may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0173] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0174] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0175] Obviously, those skilled in the art can make various modifications and variations to the embodiments of this application without departing from the scope of the embodiments of this application. Therefore, if these modifications and variations to the embodiments of this application fall within the scope of the claims of this application and their equivalents, this application also intends to include these modifications and variations. In the various embodiments of this application, unless otherwise specified or logically conflicting, the terminology and / or descriptions between the various embodiments are consistent and can be mutually referenced. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

Claims

1. A sensing method, characterized in that, Applied to a first sensor, the method includes: Acquire first data, wherein the first sensor supports different reporting modes, and the content of the first data is related to the reporting mode; Send the first data to the target node; The first data includes at least one of the following sensory data: Raw sensory data; Transceiver channel data; Point cloud level data; Feature-level data; Target-level data.

2. The method according to claim 1, characterized in that, The first data includes sensing data of different regions of interest (ROIs) within the detection area of ​​the first sensor, and the reporting modes corresponding to the different ROIs are different.

3. The method according to claim 1 or 2, characterized in that, The reporting mode of the first data is related to the scene in which the first sensor is located.

4. The method according to claim 3, characterized in that, The first sensor is installed on the vehicle, wherein, The scenario in which the first sensor is located includes any of the following scenarios based on the vehicle's driving speed: low-speed driving scenario, medium-speed driving scenario, and high-speed driving scenario; or, The scenario in which the first sensor is located includes any of the following scenarios based on the road type where the vehicle is located: urban road scenario, highway road scenario; or, The scenario in which the first sensor is located includes any of the following scenarios classified based on the meteorological conditions at the vehicle's location: good visibility scenario, moderate visibility scenario, poor visibility scenario; or, The scenario in which the first sensor is located includes any of the following scenarios based on the road conditions of the road segment where the vehicle is located: severely congested road segment scenario, moderately congested road segment scenario, lightly congested road segment scenario, and unobstructed road segment scenario.

5. The method according to claim 4, characterized in that, The reporting mode of the first data includes a combination of at least one of the following reporting modes: Raw sensing data reporting mode; Data reporting mode for transmit / receive channels; Point cloud-level data reporting mode; Feature-level data reporting mode; Target-level data reporting mode.

6. The method according to claim 5, characterized in that, Different types of sensory data meet the following priority levels: The priority of the raw sensing data is higher than the priority of the transceiver channel data; The priority of the transmit / receive channel data is higher than the priority of the point cloud-level data; The point cloud-level data has a higher priority than the feature-level data and the target-level data.

7. The method according to claim 6, characterized in that, The reporting mode corresponding to different ROIs is related to the priority of the perceived data.

8. The method according to any one of claims 4-7, characterized in that, In any of the following scenarios: low-speed driving scenario, urban road scenario, good visibility scenario, severely congested road segment scenario, or moderately congested road segment scenario, the content of the first data includes target-level data of different ROIs of the first sensor. In any of the following scenarios: medium-speed driving scenario, high-speed driving scenario, highway scenario, general visibility scenario, poor visibility scenario, lightly congested road segment scenario, or unobstructed road segment scenario, the content of the first data includes at least one of the following: raw perception data of the first ROI, transmit / receive channel data, or point cloud-level data, and target-level data of the second ROI, wherein the detection distance corresponding to the first ROI is less than the detection distance corresponding to the second ROI.

9. The method according to any one of claims 4-8, characterized in that, The method further includes: Receive first indication information from the target node; The scene in which the first sensor is located is determined based on the first indication information; The first indication information includes at least one of the following: The speed of the vehicle; The speed of other vehicles near the vehicle in question; The geographical type to which the detection area belongs; The meteorological environment information corresponding to the detection area; The load information of the vehicle's computing nodes.

10. The method according to any one of claims 2-9, characterized in that, The method further includes: The system receives a second indication from the target node, which indicates the reporting mode of the first sensor or the scene in which the first sensor is located.

11. The method according to any one of claims 2-10, characterized in that, The method further includes: Different ROIs within the detection area of ​​the first sensor are determined based on the field of view or detection distance range of the first sensor; or, Receive third indication information from the target node, the third indication information indicating different ROIs within the detection area of ​​the first sensor.

12. The method according to any one of claims 1-11, characterized in that, The target node includes a second sensor; or, The first sensor is mounted on the vehicle, and the target node includes the vehicle's computing node.

13. The method according to claim 12, characterized in that, The second sensor is located on the same vehicle as the first sensor.

14. A sensing method, characterized in that, Applied to the target node, the method includes: Receive first data from a first sensor, wherein the first sensor supports different reporting modes, and the content of the first data is related to the reporting mode; Perform perception processing based on the first data; The first data includes at least one of the following sensory data: Raw sensory data; Transceiver channel data; Point cloud level data; Feature-level data; Target-level data.

15. The method according to claim 14, characterized in that, The first data includes sensing data of different regions of interest (ROIs) within the detection area of ​​the first sensor, and the reporting modes corresponding to the different ROIs are different.

16. The method according to claim 14 or 15, characterized in that, The reporting mode of the first data is related to the scene in which the first sensor is located.

17. The method according to claim 16, characterized in that, The first sensor is mounted on the vehicle, and the target node includes either the vehicle's computing node or a second sensor associated with the vehicle. The method further includes: Sending first indication information to the first sensor, wherein the first indication information is used to determine the scene in which the first sensor is located, and the first indication information includes at least one of the following: The speed of the vehicle; The speed of other vehicles near the vehicle in question; The geographical type to which the detection area belongs; The meteorological environment information corresponding to the detection area; The load information of the vehicle's computing nodes.

18. The method according to claim 17, characterized in that, The scenario in which the first sensor is located includes any of the following scenarios based on the vehicle's driving speed: low-speed driving scenario, medium-speed driving scenario, and high-speed driving scenario; or, The scenario in which the first sensor is located includes any of the following scenarios based on the road type where the vehicle is located: urban road scenario, highway road scenario; or, The scenario in which the first sensor is located includes any of the following scenarios classified based on the meteorological conditions at the vehicle's location: good visibility scenario, moderate visibility scenario, poor visibility scenario; or, The scenario in which the first sensor is located includes any of the following scenarios based on the road conditions of the road segment where the vehicle is located: severely congested road segment scenario, moderately congested road segment scenario, lightly congested road segment scenario, and unobstructed road segment scenario.

19. The method according to claim 18, characterized in that, The reporting mode of the first data includes a combination of at least one of the following reporting modes: Raw sensing data reporting mode; Data reporting mode for transmit / receive channels; Point cloud-level data reporting mode; Feature-level data reporting mode; Target-level data reporting mode.

20. The method according to claim 19, characterized in that, Different types of sensory data meet the following priority levels: The priority of the raw sensing data is higher than the priority of the transceiver channel data; The priority of the transmit / receive channel data is higher than the priority of the point cloud-level data; The point cloud-level data has a higher priority than the feature-level data and the target-level data.

21. The method according to claim 20, characterized in that, The reporting mode corresponding to different ROIs is related to the priority of the perceived data.

22. The method according to any one of claims 18-21, characterized in that, In any of the following scenarios: low-speed driving scenario, urban road scenario, good visibility scenario, severely congested road segment scenario, or moderately congested road segment scenario, the content of the first data includes target-level data of different ROIs of the first sensor. In any of the following scenarios: medium-speed driving scenario, high-speed driving scenario, highway scenario, general visibility scenario, poor visibility scenario, lightly congested road segment scenario, or unobstructed road segment scenario, the content of the first data includes at least one of the following: raw perception data of the first ROI, transmit / receive channel data, or point cloud-level data, and target-level data of the second ROI, wherein the detection distance corresponding to the first ROI is less than the detection distance corresponding to the second ROI.

23. The method according to any one of claims 17-22, characterized in that, The first sensor is mounted on the vehicle, the target node includes the vehicle's computing node, and the method further includes: Send a second indication message to the first sensor, the second indication message indicating the reporting mode of the first sensor or indicating the scene in which the first sensor is located.

24. The method according to any one of claims 15-23, characterized in that, The method further includes: Different ROIs within the detection area of ​​the first sensor are determined based on the field of view or detection distance range of the first sensor; or, A third indication message is sent to the target node, the third indication message indicating different ROIs within the detection area of ​​the first sensor.

25. A sensing device, characterized in that, Includes units for implementing the method as described in any one of claims 1-13.

26. An electronic device, characterized in that, Includes a processor, which is coupled to memory: The processor is configured to execute a computer program or instructions stored in the memory to cause the electronic device to perform the method as described in any one of claims 1 to 13, or to cause the electronic device to perform the method as described in any one of claims 14 to 24.

27. A processing apparatus, characterized in that, Includes units for implementing the method as described in any one of claims 14 to 24.

28. A readable storage medium, characterized in that, It includes a program or instructions that, when executed, perform the method as described in any one of claims 1 to 13, or, when executed, perform the method as described in any one of claims 14 to 24.

29. A computer program product, characterized in that, When the computer program product is run on a computer, it causes the computer to perform the method as described in any one of claims 1 to 13, or causes the computer to perform the method as described in any one of claims 14 to 24.

30. A communication system, characterized in that, Includes a first sensor for implementing the method as described in any one of claims 1-13, and a target node for implementing the method as described in any one of claims 14-24.

Images